不确定性环境下数据驱动的电力系统优化调度方法综述

Overview on Data-driven Optimal Scheduling Methods of Power System in Uncertain Environment

随着可再生能源大规模并网,电力系统运行的不确定性显著增加,考虑不确定性因素的调度方法逐渐得到重视。另一方面,建立准确的不确定性因素模型是求解优化问题的前提与关键,可再生能源出力具有复杂的不确定性,其发电数据为随机调度提供了科学的数据支撑。文中总结了不确定性环境下数据驱动的电力系统调度的理论方法以及应用场景。首先,总结了传统随机优化调度中数据驱动的随机变量建模方法。其次,介绍了鲁棒优化调度中数据驱动的不确定性集合建模方法。然后,针对随机优化中不确定性因素建模不准确以及鲁棒优化结果较为保守的问题,重点阐述了基于数据驱动分布鲁棒的电力系统优化调度理论与方法,梳理了随机变量的概率分布模糊集构建方法和分布鲁棒优化的模型构建及求解算法。最后,对数据驱动的电力系统调度未来的研究工作进行了展望。

With the large-scale integration of renewable energy,uncertainties significantly rise in power system operation.Scheduling methods that take uncertainties into account gradually draw more and more attentions.Meanwhile,the establishment of an accurate uncertain model is the premise and key to solve the optimization problem.The output of the renewable energy has complex uncertainties,and its power generation data provide scientific data support for stochastic scheduling.This paper summarizes the theoretical methods and application scenarios of data-driven scheduling for power systems in uncertain environments.Firstly,data-driven modeling methods of random variables in traditional stochastic optimal scheduling are summarized.Secondly,modeling methods of data-driven uncertain sets in robust optimal scheduling are introduced.Then,aiming at the problems of inaccurate modeling of uncertain factors and conservative results of the robust optimization in stochastic optimization,the theory and method of optimal scheduling for power systems are explained based on data-driven distributed robustness;and the construction method of probability distribution fuzzy set of random variables,the model construction and solution algorithm of distribution robust optimization are sorted out.Finally,the future research work of data-driven scheduling for power systems is prospected.